Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Chapter Five :Population Growth and the Environmental Crisis: Beyond the 'Obvious'

Franck Amalric

On many accounts, the twentieth century has witnessed an unprecedented transformation of the world. Economic growth, technological innovation, the shrinking of the world due to the emergence of modern means of transport and of communication, the globalization of the economy, sometimes of politics, two world wars and the multiplication of regional conflicts, are among the many phenomena of historical dimension that have reshaped the world during this century. And undeniably, the increase in the number of persons living on Earth is another facet of this profound transformation.

The figures are indeed impressive: at the turn of the century, world population is estimated to have been about 1.5 billion. It reached 2 billion around 1930, 2.5 billion in 1950. It is now well over 5 billion, and prospects for the year 2025 are between 7.6 and 9.4 billion, with a medium variant at 8.5 billion(UN 1991). Although such an increase necessarily raises a number of issues, in particular the capacity of adaptation of the different societies in which it takes place, it is still necessary to stress that the number of people living on earth cannot be a problem per se, but only the possible cause of a problem. Health, particularly of women, the environmental crisis, poverty, food security, international security, are the ultimate issues that need to be addressed. It is only in relation to them that looking at the consequences of population growth becomes relevant.

As in the rest of this volume, this chapter is mainly concerned with issues pertaining to the global level. It therefore leaves out a number of topics often associated with population growth, like the impact on the local environment or the consequences for economic development. Our focus here is population growth in the perspective of the global environmental crisis.

The contribution of population growth to the environmental crisis is a very controversial topic. In the recent literature, however, there has been several attempts to synthesize the different views within a single theoretical framework, known as the Ehrlich equation. This paper shows the limits of this framework, not only in its capacity to explain the process of environmental degradation, but also in its capacity to provide guidelines for policies. Two shortcomings of the equation are discussed. First, although it is presented as giving a neutral account of the issue, in practice it appears decisively biased against the population factor. Second, applying the equation can only assess the contribution of one factor in generating additional environmental degradation, leaving out those factors that created the problem in the first place. In response, I propose an alternative way to use the equation, one which yields quite different results indeed. But let us start by giving a short account of the debates around population growth.


As one facet of the transformation of the world, population growth occupies a special place. First, although undeniably a central feature of this century, it has nevertheless been opposed to the larger historical trend, that of modernization. The world has been viewed as a world in transition towards development, and population growth as stemming from behaviours - namely high fertility rates - characterizing the non-modern or undeveloped world. Thus, the prevailing theory of fertility of these last thirty years, the so-called demographic transition theory, associated a decline of fertility rates with the process of modernization. The view was taken up by policy-makers around the developing world, as epitomized by Indira Gandhi's claim, made in 1972 as Prime Minister of India, that 'development is the best contraceptive',2 or as defended by the developing countries at the 1974 World Conference on Population in Bucharest.

Second, it underlines the split between developed and developing countries, and is therefore a highly political topic. Since 1960, almost 90 per cent of world population growth occurred in developing countries, and the trend will be even more accentuated in the future decades (UN 1991). As a potential factor of environmental degradation, population growth becomes therefore a topic of international negotiations. But even without referring to the global environmental crisis, population growth in the South is seen as a destabilizing factor in the world. At the extreme, it is presented as the cause of all ills. Consider, for instance, the following declaration made in 1988 by the Club of Earth, whose members all belong to both the US National Academy of Sciences and the American Academy of Arts and Sciences:

Arresting population growth should be second in importance only to avoiding nuclear war on humanity's agenda. Overpopulation and rapid population growth are intimately connected with most aspects of the current human predicament, including rapid depletion of non-renewable resources, deterioration of the environment (including rapid climate change), and increasing international tensions.

Cited in Ehrlich and Ehrlich 1990:18

This vision of the issue by the scientific community is not isolated. The famous Heidelberg Declaration signed by 425 prominent scientists, including 52 Nobel prize laureates, at the time of the end of the conference in Rio called overpopulation a 'plague' comparable to 'hunger and pandemics'.

This vision is, in part, purely egoistic. In Rufin's words it is the fear of the 'new barbarians' (Rufin 1989), that a degradation of the situation in the South will threaten the North's welfare. As Lorimer, a leading demographer, bluntly put it:

trends which threaten the national aspirations of more than half the world's population present a problem to all nations. Frustration breeds envy, suspicion and violence. The security of the lucky nations with large national resources, accumulated wealth and advanced techniques may be critically affected by the progress or reverse experienced in the less fortunate nations during the next few decades.

Lorimer 1963: 145

In response, Southern countries have imposed the view that Northern lifestyles, and not Southern population growth, was the central cause of the global environmental crisis (see the chapter in this book by Lipietz). The absence of any official debate on population growth in Rio - although the topic was on everybody's lips in the corridors - can be seen as a diplomatic victory of the South... and of the Vatican.

But the North-South divide is not that clear, because population growth also underlines the split between the westernized elites of the South and the vast majority of the people living in those countries. To have a large number of children is often the best strategy for the poor: children are rapidly productive; they are an insurance for old age; indeed, each child is an additional chance of economic success for the family. But for the governing bodies of developing countries, population growth is increasingly perceived as hindering economic development. It raises the demand for capital so as to increase the capital available per worker; it increases the demand for social services, notably in education, and thereby diverts some resources from more directly productive investments (see Lewis 1955; Coale and Hoover 1958). Population policies are thus designed in the perspective of a race between population growth and economic development. Note that the long-term effects of population growth notably population density - are given little attention (Stamper 1977). Today, many developing countries have official population planning policies geared at slowing down the rate of population growth. These include China, the nations of South Asia, Central America, South-East Asia and parts of Africa. Conversely, most states of South America, West Asia and some of Africa do not have any population programme (UN 1990). Altogether, about 61 per cent of people in the world live under a government which incites them to have fewer children. The North - from the point of view of sustainability - and the westernized elites of the South - from the point of view of development - could thus find a common interest in controlling population growth from the perspective of 'sustainable development'. Already in the 1984 World Conference on Population in Mexico, the South was asking for financial aid from the North to run population planning policies. In these conditions, who will protect the basic reproductive rights of the poor ?

For these different reasons, population growth may seem a very convenient and (statistically) highly visible cause of environmental degradation. Because it stems from non-modern behaviours, it does question the modern ideology. Furthermore, it appeals to common sense: environmentalists consider that the level of human activities on earth is reaching unsustainable limits. Since the total of human activities is the sum of each person's activity, it would appear intuitively correct to say that the more persons there are, the greater the total activity. Thus population growth is often held, notably by biologists and international organizations, as an important factor, if not the main one, of environmental degradation (Ehrlich and Ehrlich 1990; WCED 1987; UNEP 1990; Myers 1991).

But the reality is much more complex because different people have varying impacts on the environment. Lifestyles is the other main possible factor of environmental degradation. If sustainability puts some constraints on the total level of human activity, it does not, at least initially, determine how this activity is to be distributed. In theory, a sustainable society could equally be characterized by a small number of people with a high level of consumption per person, or by a large number of people consuming little resources but enjoying long periods of leisure. In this respect, a number of authors have argued that modern institutions are the real culprit, and that population growth has only a marginal impact on the environment. Inegalitarian market relations would displace subsistence agriculture for export crops nourishing the luxury demands of the rich (Lappé and Collins 1980). Indeed a number of recent famines have been explained, not as an absolute scarcity in food, but as a scarcity of entitlements to food (See, 1981). Environmental problems linked to agriculture - desertification, deforestation, erosion of soils due to over use, pollution due to the over use of pesticides and fertilizers - would thus stem from a misuse of land under the pressure of market forces. Another important factor would be modern technology, oriented toward economic efficiency at the cost of environmental degradation (Commoner 1971).


In response to this chorus of conflicting views, there has been a number of attempts to arrive at a synthesis and to measure the contribution of each factor. This required a theoretical framework, often referred to as the 'Ehrlich equation'. The equation combines total impact on the environment, population and consumption in the following way:
total impact = population x (total consumption/population) x (total
impact/total consumption) [1]

To simplify, we shall henceforth write:

I = PxC'xT.

C' is consumption per capita and must be distinguished from C, total consumption.

This equation is merely an identity and is therefore always valid. In fact, C' and T are never observed directly. I, T. and C are the independent variables, in the sense that we can usually get direct data for them, and straightforwardly

C'= C/P, T = I/C [2]

The goals of the equation are:

The value of the equation must be gauged by its capacity to meet these two goals, the former of which is dealt with in this section, the latter one in the next section.

We cannot derive directly from the Ehrlich equation any indication as to the contribution of population growth to the total environmental impact. What is usually done is to compare the levels of impact at two different points in time; what is then assessed is the contribution of population growth to the increase in the total impact. But to do so. it is necessary to make three assumptions. First, to choose one definition, among many possible, of such contribution. Second, to consider the groups as homogenous in terms of consumption and technology. Third, to suppose that the variables P. T and C' are independent. These assumptions are repeatedly made in the literature without much explanation or empirical support. Yet they are problematic in many ways, as we shall now see.

What Definition ?

To examine this issue, we can use a simplified version of equation [2]4
total impact = population x impact per person, or I = P x F [3]

With this equation, there is a number of possible definitions of the contribution of population growth to a change in total impact. This is merely a new version of the old index-number problem which arises each time we attempt to split changes up into contributions. A simple numerical example will make the point:

Date 1: P=100 I=100 F=1
Date 2: P=200 I=400 F=2

Total increase of environmental impact is therefore 300. There are two extreme methods of defining the contribution of population growth. One is to look at what would have been the impact of population growth if there had not been any increase in the level of impact per person. Under this hypothesis, the contribution would have been 100. Thus the contribution of population growth would be 33 per cent. Another way, rarely used, is to take the increase in impact per person as a given, and look at what would have been the impact had there not been any population growth. In this case, the contribution of population growth is assessed at 200 or 66 per cent of the total.

Perhaps the more commonly used method is to define this contribution as the ratio of the rate of population growth to the rate of growth of the environmental impact. In our two period example, this third method is equivalent to the first one. But to use rates of growth is itself not very satisfactory because the result will depend on the number of sub-periods considered. It is theoretically more appropriate to suppose that there is an infinity of sub-periods. In this case, the contribution will be determined by the logarithms of the variables. This definition has the important advantage of being independent of the initial level of impact per person (Fo).

We have therefore three competing definitions (assuming that P. F and I move in the same direction):

1 sp = (P2-P1)/(I2-I1) X F1
2 Sp = (P2-P1)/(I2-I1) x F2
3 Sp = (Log P2-Log P1)/(Log I2-Log I1)
where Sp stands for the 'share of population growth'.

Still others could be proposed, and each definition will yield very different results. Thus, in our numerical example, we get: Sp (1) = 33 %; Sp (2) = 67 %; Sp (3) = 50 %. The variability of the results is illustrated in the case of world carbon dioxide emissions between 1960 and 1988 in Tables 5.1.

Table 5.1 Regional carbon dioxide emissions, 1960 1988

impact %
(3) %
1960 1988 1 2 3
Africa 40 170 279 605 36 70 53 2
Asia (dev) 297 1193 1574 2876 27 60 43 12
Latin America 82 267 218 430 43 71 57 3
North America 852 1430 199 272 54 66 60 11
Europe 732 1180 425 496 27 38 32 5
USSR 396 1086 214 284 19 39 28 6
Japan 64 270 94 122 9 28 18 1
Oceania 28 75 16 26 37 61 49 1
Developing countries 418 1630 2072 3911 31 63 46 17
Developed countries 2072 4040 948 1200 28 43 35 22
World 2490 5670 3020 5111 [29] [51] [41]  

Source: Harrison (1992).

Notes: 3 corresponds to Harrison's results. Contributions at the world level are calculated
indirectly, weighted by increases in emissions.

The last column is the share of population growth to total world increase in emissions
using definition 3.

Definition 1 (respectively 2) considers implicitly Fit (respectively F2) as the 'normal' level of impact per person. The difference between 1 and 2 stems therefore from a difference in what level of consumption per person is taken as a reference. Definition 3, by contrast, assesses the impact of population growth as dependent on the impact level when each bit of the population growth occurred. It is therefore independent of any fixed level of reference of impact per person. But it does not mean that it is free of any reference. In fact, its reference is dynamic. Under definition 1, each 'additional' person has a 'right' to an impact of F1. Under definition 3, each additional person has a right to an impact which corresponds to the average impact when the person comes into the world.

By a similar argument, but considering now the contribution of the growth in consumption per person with the constraint that the sum of the two contributions must be equal to one or 100 per cent, we derive that the different definitions also take implicitly as a reference a certain size of population: P2 for definition 1, P1 for definition 2, and a dynamic reference
for definition 3. Saying, with definition 1, that population growth accounts for 33 per cent of the increase of total impact means that, if there had not been any growth in impact per person, then total impact would have risen only by 33 per cent.

There is no overriding reason to favour one definition over the other. Definition 3 has indeed the advantage of being independent of any a priori choice of a level of reference. Yet this is a choice in itself. And if it is mathematically convenient, it should however be supported on other grounds. The point is that it is impossible to measure the 'contribution' of population growth to the environmental impact of human activities without adopting references for both population size and consumption per person. And different references will yield very different results.


Contents - Previous - Next